Flow Efficiency
Flow efficiency compares active working time with total elapsed flow time to show how much of an item's journey is spent being worked rather than waiting.
Download Markdown (.md)Soroush’s Product Wiki · https://wiki.soroush.uk/topic/flow-efficiency/
01What it means
Flow efficiency is the proportion of total elapsed flow time during which an item is actively worked. A common calculation divides active time by total elapsed time and expresses the result as a percentage. Teams must define what counts as active work, waiting, and the measurement boundary before the number has meaning.[1][2]
- Start boundary
- When elapsed flow time begins
- Active time
- States counted as work
- Waiting time
- Queues, handoffs, blocked states
- Finish boundary
- When the item leaves the measured flow
Calculate per item before summarizing; the result only has meaning when population, boundaries, and state definitions are explicit.
02Why it matters
The measure challenges the intuition that slow delivery is mainly caused by people working too slowly. It can reveal that queues, handoffs, blocked states, approvals, or batching dominate elapsed time. This redirects improvement toward removing avoidable waiting and redesigning the system rather than demanding greater individual utilization.
03When to use it
Use flow efficiency when detailed state or timestamp data is available and a specific delay deserves investigation. It is most useful as a diagnostic for comparable item classes or workflow segments, especially alongside cycle time, WIP, aging, and qualitative observation. Measure a sample manually before automating an ambiguous definition.
04How it works
- Choose the item population, start and finish boundary, and a precise definition of active versus waiting time.
- Capture time in each state or reconstruct it from reliable timestamps for completed items.
- Calculate active time divided by total elapsed time for each item before summarizing the distribution.
- Inspect where waiting accumulates and validate the classification with people who perform the work.
- Test changes such as smaller batches, earlier review, fewer handoffs, or WIP controls, then reassess.
05A practical example
A compliance change takes 20 elapsed days but receives active analysis, design, implementation, and review for only five. Its flow efficiency is 25 percent under the team's definition. Investigation finds repeated approval queues, so the group schedules earlier compliance pairing and measures whether waiting and total cycle time fall together.
06Where it breaks
Active and waiting categories are often subjective, and digital tools may confuse status with actual attention. A higher percentage is not automatically better: necessary incubation, customer response time, or risk controls can create legitimate waiting. Teams can game the metric by relabeling states, and local efficiency may harm overall outcomes.
Common mistakes
- Treating every non-active state as waste without considering necessary feedback, risk, or customer timing.
- Reporting one aggregate percentage that hides variation among item types and workflow segments.
- Optimizing the ratio while total cycle time, quality, or delivered value becomes worse.
07Sources
- Open Guide to Kanban (July 2025)Kanban Guides · Open professional guideOpen Guide to Kanban (July 2025)
- Flow Metrics Work Items View Dashboard TemplateAtlassian Support · Product documentationFlow Metrics Work Items View Dashboard Template